Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A convolutional neural network

A convolutional neural network, convolutional technology, applied in the field of convolutional neural networks

Active Publication Date: 2018-10-23
FOTONATION LTD
View PDF18 Cites 15 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, it should be appreciated that the architecture chosen for training may need to be iteratively tuned to optimize the classification provided by the CNN

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A convolutional neural network
  • A convolutional neural network
  • A convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0041] image 3 A block diagram of a CNN engine 30 implemented in an image acquisition system according to an embodiment of the present invention is shown. CNN engine 30 is connected to system bus 42 and has access to main (DRAM) memory 40 into which images acquired by the system are written. The image acquisition pipeline for obtaining and possibly preprocessing images before or after writing them to DRAM 40 is well known and will not be described in detail here, but is described in the above-mentioned U.S. Provisional Application No. 26, 2015. Examples of such systems are described in 62 / 210,243 (Ref: FN-469), PCT applications WO2014 / 005783 (Ref: FN-384) and US2015 / 262344 (Ref: FN-384-CIP).

[0042] Accordingly, an application program (not shown) executed by system CPU 50 may signal to controller 60 within CNN engine 30 via system bus 42 a region of interest (ROI) of an image stored in DRAM 40 and indicated by CPU 50. ) will be analyzed and categorized. The controller 60 ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

A convolutional neural network (CNN) for an image processing system comprises an image cache responsive to a request to read a block of N*M pixels extending from a specified location within an input map to provide a block of N*M pixels at an output port. A convolution engine reads blocks of pixels from the output port, combines blocks of pixels with a corresponding set of weights to provide a product, and subjects the product to an activation function to provide an output pixel value. The image cache comprises a plurality of interleaved memories capable of simultaneously providing the N*M pixels at the output port in a single clock cycle. A controller provides a set of weights to the convolution engine before processing an input map, causes the convolution engine to scan across the input map by incrementing a specified location for successive blocks of pixels and generates an output map within the image cache by writing output pixel values to successive locations within the image cache.

Description

technical field [0001] The present invention relates to a convolutional neural network with improved image caching. Background technique [0002] The processing flow of a typical convolutional neural network (CNN) is as follows: figure 1 shown. Typically, the input to a CNN is at least one 2D image / map 10 corresponding to a region of interest (ROI) from an image. The image / map may include only image intensity values, e.g., from the Y plane of a YCC image; or the image / map may include any combination of color planes from the image; or alternatively or in addition, the image / map may contain values ​​from Image derived values ​​such as the Histogram of Gradients (HOG) map described in PCT Application No. PCT / EP2015 / 073058 (Ref: FN-398), the disclosure of which is incorporated herein by reference, or the integral image map. [0003] CNN processing consists of two stages: [0004] ○ Feature extraction (12) - convolution part; and [0005] o Feature Classification (14). [0...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G06K9/46G06N3/06
CPCG06N3/063G06N3/082G06F12/0851G06F2212/454G06F2212/455G06V10/955G06V10/454H04N19/167H04N19/17G06N3/045
Inventor P·比吉奥伊M·C·蒙特亚努A·卡里曼C·扎哈里亚D·迪努
Owner FOTONATION LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com